A method for fault-tolerant control and interference utilization of drone swarm

Abstract

The invention discloses a method for fault-tolerant control and interference utilization of unmanned aerial vehicle clusters based on a switching function method, and belongs to the technical field of coordinated control of unmanned aerial vehicle clusters. The invention comprises: establishing a dynamic model of an unmanned aerial vehicle attitude system with unknown time-varying fault direction; designing an estimator to estimate the attitude angle output of a leader and generate a reference signal; designing a working mode and a switching function of the unmanned aerial vehicle attitude system; designing an interference observer and an interference effectiveness evaluation index for mismatched interference; designing input soft saturation and an ideal control signal for the unmanned aerial vehicle attitude system; the invention realizes fault-tolerant control of unmanned aerial vehicle clusters while ensuring the transient performance of unmanned aerial vehicles after reverse faults, and further accelerates the convergence speed and saves energy consumption by suppressing harmful interference and utilizing beneficial interference.

Description

Technical Field

[0001] The present invention relates to the technical field of cooperative control of unmanned aerial vehicle clusters, and specifically to a method for fault-tolerant control and interference utilization of unmanned aerial vehicle clusters. Background Art

[0002] With the advancement of science and technology, cooperative control of unmanned aerial vehicles has received extensive attention in recent years. With the characteristics of high efficiency, high flexibility and strong robustness, it has shown great application value in various fields. As an important branch of cooperative control, swarm control (group consistency control) refers to the cooperation of multiple unmanned aerial vehicles to complete multiple control objectives, which is manifested in that the motion postures of the leaders of each sub-group are different, while the unmanned aerial vehicles in the same sub-group have the same leader. Since the output of the unmanned aerial vehicle attitude system is determined by the rotor speed of the unmanned aerial vehicle, when a motor failure occurs in the unmanned aerial vehicle, it may manifest as a reverse fault. Compared with efficiency loss fault and bias fault, reverse fault is more harmful to the stable operation of the unmanned aerial vehicle. Although the existing fault-tolerant method based on Nussbaum function ensures the bounded stability of the system under reverse fault, the transient performance of the system is poor after the reverse fault occurs, which is specifically manifested in that the input signal of the system has a reverse pulse of uncertain duration after the reverse fault occurs and before the Nussbaum function changes its sign. Since there is a mutually reinforcing loop between the reverse pulse and the system error, the growth rate of the reverse input is getting faster and faster, and excessive reverse pulses may cause irreversible damage to the system. In addition, after the Nussbaum function changes its sign, the input of the system has a large amplitude pulse due to the jump of the Nussbaum function sign, which causes the system state to jitter. This is determined by the characteristics of the Nussbaum function method itself. Therefore, it is necessary to propose a new fault-tolerant control method for reverse faults to prevent excessive reverse input from causing irreversible damage to the system while suppressing the input peak phenomenon and state jitter phenomenon that occur after the control gain changes its sign.

[0003] On the other hand, due to the complexity of the UAV working environment and the uncertainty of system modeling, UAVs will be affected by interference in both matched and unmatched channels. Most of the existing interference suppression methods only consider the harmful effects of interference on system performance, while ignoring the implicit beneficial characteristics of interference. In fact, by making reasonable use of interference, the convergence speed of the system can be effectively accelerated, the steady-state error of the system can be reduced, and the energy consumption of the system can be reduced. Therefore, by analyzing the characteristics of interference, making reasonable use of beneficial interference and compensating for harmful interference, it is very helpful to improve system performance. Based on the above analysis, in order to ensure the steady-state performance and transient performance of UAVs under interference and unknown direction faults, the present invention proposes a UAV cluster grouping fault-tolerant control and interference utilization method based on the switching function method to improve system performance and reduce energy consumption. Summary of the invention

[0004] The purpose of the present invention is to propose a method for fault-tolerant control and interference utilization of UAV clustering to solve the problems of poor transient performance of the existing Nussbaum function method after an unknown direction fault occurs and the situation where beneficial interference is not considered. The present invention improves the transient performance of the UAV cluster attitude system after a reverse fault occurs, and reduces the energy consumption of the UAV and accelerates the convergence speed by utilizing beneficial interference and compensating for harmful interference.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for fault-tolerant control and interference utilization of drone clustering includes the following steps:

[0007] S1. Establish a dynamic model of the UAV attitude system with unknown time-varying fault direction;

[0008] S2, designing an estimator to estimate the attitude angle output of the leader and generate a reference signal;

[0009] S3. Design the working mode and switching function of the UAV attitude system;

[0010] S4. Design interference observer and interference effectiveness evaluation index for mismatched interference;

[0011] S5. Design the ideal control signal for input soft saturation and UAV attitude system.

[0012] Preferably, the UAV attitude system in S1 includes a follower UAV attitude system with unknown time-varying fault direction and a leader UAV attitude system with unknown time-varying fault direction;

[0013] The function of the attitude system of the following UAV with unknown time-varying fault direction is expressed as:

[0014]

[0015] y i，p =x i，p，1 , i=1,...,N

[0016] Among them, x i，1,1 、x i,2,1 、x i，3,1 Respectively represent the roll angle, pitch angle and yaw angle of the UAV attitude system; x i，1,2 、x i，2,2 、x i，3,2 Represent the angular velocities of roll, pitch and yaw respectively; represents the state vector; d i，p，1 and d i，p，2 denote the unmatched disturbance and matched disturbance with bounded derivatives respectively; y i，pRepresents the Euler angle output of the drone; f i，p A nonlinear term representing the attitude of the drone; is the torque under the UAV fault, which is expressed as the system input:

[0017]

[0018] Among them, u i，p represents normal torque input; ρ i，p represents the failure factor, satisfying ρ i，p ≠0; r i，p Indicates additive fault;

[0019] The nonlinear term of the above UAV attitude is specifically expressed as:

[0020]

[0021] Among them, J x , J y , J z , J r Represents the moment of inertia and rotor inertia in the x, y, and z directions of its own coordinates; I p Indicates air resistance parameters, p = 1, 2, 3; Ω r represents the differential angular velocity of the rotor; the communication topology of the followers in the drone cluster contains a directed spanning tree and the root node can obtain the signal of the leader;

[0022] The function of the leader UAV attitude system with unknown time-varying fault direction is expressed as:

[0023]

[0024] in, and They represent the leader’s Euler angle and the corresponding angular velocity respectively; l i Indicates the number of the group to which drone i belongs; They represent the output, input and nonlinear term of the leader UAV respectively; the nonlinear term of the leader UAV has the same form as that of the follower UAV;

[0025] Assume that the derivative of the leader UAV output is bounded, expressed as in Is a normal number.

[0026] Preferably, S2 specifically includes the following contents:

[0027] The communication interaction between drones is used to design an estimator for each follower drone to estimate the leader's information and achieve group consistency of the output signals of each estimator. The specific function is expressed as:

[0028]

[0029] in, represents the output of the estimator; Represents the input signal of the observer, and its function is expressed as:

[0030]

[0031] Where c represents a positive constant; represents the output vector of the leader drone and its derivative is ω i represents the group consistency error; and Respectively represent group l i The set of drones that have signal transmission and reception links with other drones in the group; represents the neighbor set of drone i; b i Represents the traction gain, which has b for the traction node i =1, otherwise b i =0; follow the root node and collection of the drone topology graph and The nodes in a are pulling nodes; ij with a ik Represents the topological weight between UAVs.

[0032] Preferably, S3 specifically includes the following contents:

[0033] According to the dynamic model and fault model of the UAV attitude system, there are 8 potential fault direction modes. The designed working mode set encoded in binary mode with {-1,1} as elements is:

[0034]

[0035] Where diag{·} represents a diagonal matrix;

[0036] Based on the above working mode, the following switching function is designed for each drone:

[0037]

[0038] in, Indicates the working mode of drone i; bin2dec indicates binary to decimal conversion; sign indicates the sign function; α i (t) and Represents auxiliary functions; is the event triggering moment, and its judgment conditions are:

[0039]

[0040] Among them, the event condition g i,p (t) is calculated as:

[0041] g i,p (t)=|z i,p (t)|-β(t)

[0042] Among them, β(t) is the set event monitoring boundary, and the function is expressed as:

[0043]

[0044] Among them, ε0, ε ∞ and is a positive constant and satisfies ε0>ε ∞ .

[0045] Preferably, the S4 specifically includes the following contents:

[0046] Design a disturbance observer, specifically:

[0047]

[0048] in, represents the observed output of the mismatched disturbance, ψ i,p,1 represents the auxiliary function, φ i,p,1 >1 is a constant;

[0049] Based on the above interference observer design, the interference effectiveness evaluation index is further designed. The specific contents are as follows:

[0050]

[0051] Among them, θ i,p,1 Represents the interference effectiveness evaluation index, when θ i,p,1 >0, it indicates that the disturbance is harmful to the completion of the control target and full compensation is performed; when θ i,p,1 When <0, it indicates that the interference is beneficial to the completion of the control target, and faster convergence and lower energy consumption can be achieved by utilizing the interference; e i,p1 It represents the conversion value of the adjustment error, and its calculation expression is:

[0052]

[0053] Among them, z i,p Indicates adjustment error; and k i,p They represent the conversion function and the safety distance respectively, and their calculation expressions are:

[0054]

[0055] in, δ and a i,p is a positive constant, δ≥2.

[0056] Preferably, the S5 specifically includes the following contents:

[0057] Based on the working mode and switching function designed in S3, the input signal of the drone is expressed as:

[0058]

[0059] in, It represents the saturation input implemented at the software level, and its calculation expression is:

[0060]

[0061] Among them, u sat represents a normal number, Represents the ideal input signal of the design, and its calculation expressions are respectively:

[0062] 1) No fault condition, at this time s i,p =1, p = 1, 2, 3; the ideal input signal design is:

[0063]

[0064] Among them, c i,p,2 Constant, c i,p,2 >1; Z i =[x i,1,1 ,x i,1,2 ,x i,2,1 ,x i,2,2 ,x i,3,1 ,x i,3,2 ] T represents the input of the neural network; Γ(Z i，p ) represents the Gaussian kernel function; Indicates the adjustment error of the drone’s angular velocity; represents the output signal of the filter; represents the adaptive parameter, The update law and interference effectiveness evaluation function h(θ i,p,2 ) and auxiliary function μ i,p1 The calculation expression is:

[0065]

[0066] Among them, r i,p1 、r i,p2 and l i is a positive constant; θ i,p,2It represents the interference effectiveness evaluation index of unknown aggregate interference, and its calculation method is:

[0067]

[0068] The filter design for an ideal input signal is:

[0069]

[0070] Among them, v i,p is a positive constant; α i,p (t) is the designed virtual control law, and its calculation expression is:

[0071]

[0072] Among them, c i,p,1 >1 is a positive constant; h(·) represents the interference effectiveness evaluation function; μ i,p2 Represents an auxiliary function, whose calculation expression is:

[0073]

[0074] 2) After switching the working mode, s i,p =0; the ideal input signal and virtual control law are designed as follows:

[0075]

[0076] The meaning and calculation formula of each symbol in the above formula are the same as the corresponding symbols in step 1).

[0077] Compared with the prior art, the present invention provides a method for fault-tolerant control and interference utilization of drone clustering, which has the following beneficial effects:

[0078] (1) This invention proposes for the first time a fault-tolerant control method based on the switching function method to solve the problem of unknown time-varying fault direction in UAV clusters; based on the proposed new switching function, the UAV can directly switch to the desired working mode.

[0079] (2) This method solves the problem of excessive reverse input and system state jitter after a reverse fault occurs in the existing Nussbaum function method by switching functions and saturation functions, thereby improving the transient performance of the UAV after a reverse fault occurs.

[0080] (3) In the present invention, by designing an interference observer and an interference effectiveness evaluation index, the utilization of beneficial interference and the compensation of harmful interference are realized, which can not only complete the control target of the UAV faster, but also reduce the energy consumption of the UAV. In addition, interference utilization reduces the impact of soft saturation limitation on system performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 This is a schematic diagram of the overall process of the method for fault-tolerant control and interference utilization of drone clustering mentioned in Example 1 of the present invention;

[0082] Figure 2 This is a control principle diagram of the UAV cluster grouping fault-tolerant control and interference utilization method mentioned in Example 1 of the present invention. DETAILED DESCRIPTION

[0083] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention are described in further detail below.

[0084] Embodiment 1:

[0085] See also Figure 1-2 The present invention provides a method for fault-tolerant control and interference utilization of unmanned aerial vehicle clusters, the method comprising the following steps:

[0086] S101, establish a dynamic model of the UAV attitude system with unknown time-varying fault direction, specifically including the following contents:

[0087] The dynamic model of the attitude system of the following UAV with unknown time-varying fault direction is:

[0088]

[0089] Among them, x i,1,1 、x i,2,1 、x i,3,1 Respectively represent the roll angle, pitch angle and yaw angle of the UAV attitude system. i,1,2 、x i,2,2 、x i,3,2 are the angular velocities of roll, pitch and yaw respectively.

[0090] is the state vector, d i,p,1 and d i,p,2 denote the unmatched disturbance and matched disturbance with bounded derivatives, y i,p is the Euler angle output of the drone; f i,p is the nonlinear term of the UAV attitude, is the torque under the UAV fault, which is expressed as the system input:

[0091]

[0092] Among them, u i,p represents the normal torque input, ρ i,p The failure factor satisfies ρ i,p ≠0; r i,p is an additive fault. The nonlinear term of the drone attitude is expressed as:

[0093]

[0094] Among them, J x , J y , J z , J r is the moment of inertia in the x, y, and z directions and the rotor inertia in its own coordinates, I p Indicates air resistance parameters, p = 1, 2, 3, Ω r is the differential angular velocity of the rotor. In addition, the communication topology of the followers in the drone cluster contains a directed spanning tree and the root node can obtain the signal of the leader.

[0095] Similar to the follower drone, the leader drone attitude system is expressed as:

[0096]

[0097] in, and are the leader’s Euler angle and corresponding angular velocity, l i is the number of the group to which drone i belongs, are the output, input and nonlinear term of the leader UAV, respectively. The nonlinear term has the same form as that of the follower UAV. Here, it is assumed that the derivative of the output of the leader UAV is bounded, expressed as in is a positive constant;

[0098] S102, designing an estimator to estimate the attitude angle output of the leader and generate a reference signal, specifically including the following contents:

[0099] The communication interaction between UAVs is utilized to design an estimator for each follower UAV to estimate the leader's information and achieve group consistency of the output signals of each estimator.

[0100]

[0101] in, is the output of the estimator, is the input signal of the observer, which can be expressed as:

[0102]

[0103] Where c is a positive constant; is the output vector of the leader drone and its derivative is ω i is the group consistency error; and Respectively represent group l iThe set of drones that have signal transmission and reception links with other drones in the group. is the neighbor set of drone i; b i is the pulling gain, and there is b for the pulling node i =1, otherwise b i = 0. The root node and set of the topological graph of the drone in this patent and The nodes in a are pulling nodes; ij with a ik is the topological weight between UAVs.

[0104] Defining the estimation error Combining (5) and (6) we have:

[0105]

[0106] Among them, η i is an auxiliary variable. Further, its compact form is

[0107]

[0108] in, Similarly, the compact form of (7) is expressed as

[0109]

[0110] Consider the following Lyapunov function:

[0111]

[0112] Among them, Θ is a positive definite matrix. Taking its derivative, we can get

[0113]

[0114] in, is a positive definite matrix. Therefore, from (12), the designed observer can estimate the output signal of the leader UAV and achieve group consistency. ;

[0115] S103, design the working mode and switching function of the drone attitude system, which specifically includes the following contents:

[0116] According to the dynamic model and fault model of the UAV attitude system, there are 8 potential fault direction modes. The designed working mode set encoded in binary mode with {-1,1} as elements is:

[0117]

[0118] Where diag{·} represents a diagonal matrix.

[0119] Based on the above working mode, the following switching function is designed for each drone

[0120]

[0121] in, is the working mode of drone i, bin2dec is the operation of converting binary to decimal, and sign is the sign function. i (t) and For auxiliary functions. is the event triggering moment, and its judgment condition is

[0122]

[0123] Among them, the event condition g i,p (t) is calculated as:

[0124] g i,p (t)=|z i,p (t)|-β(t) (16)

[0125] Among them, β(t) is the set event monitoring boundary, which can be expressed as:

[0126]

[0127] Among them, ε0, ε ∞ and is a positive constant and satisfies ε0>ε ∞ ;

[0128] S104, designing an interference observer and interference effectiveness evaluation index for mismatched interference, specifically including the following contents:

[0129] The disturbance observer is designed as:

[0130]

[0131] in, is the observed output of the mismatched interference, ψ i,p,1 represents the auxiliary function, φ i,p,1 >1 is a constant. The designed disturbance observer satisfies the following conditions:

[0132]

[0133] in, is the estimated error of the interference.

[0134] Based on the above interference observer, the interference effectiveness evaluation index is designed as follows:

[0135]

[0136] Among them, θ i,p,1 is the interference effectiveness index. When it is greater than 0, it indicates that the interference is harmful and needs to be fully compensated. When it is less than 0, it indicates that the interference is beneficial to the completion of the control target, and faster convergence and lower energy consumption can be achieved by utilizing the interference. i,p1 To adjust the error conversion value, the calculation expression is:

[0137]

[0138] Among them, z i,p To adjust the error, and k i,p They are conversion function and safety distance respectively, and their calculation expressions are:

[0139]

[0140] in, δ and a i,p is a positive constant, δ≥2;

[0141] S105, design the ideal control signal for input soft saturation and UAV attitude system, including the following contents:

[0142] Based on the working mode and switching function designed in S103, the input signal of the drone is expressed as:

[0143]

[0144] in, is the saturation input implemented at the software level, and its calculation expression is:

[0145]

[0146] Among them, u sat represents a normal number, is the ideal input signal designed, and its calculation expressions are respectively:

[0147] 1) No fault condition, at this time s i,p =1,p=1,2,3. The ideal input signal design is:

[0148]

[0149] Among them, c i,p,2 Constant, c i,p,2 >1,Z i =[x i,1,1 ,x i,1,2 ,x i,2,1 ,x i,2,2 ,x i,3,1 ,xi,3,2 ] T is the input of the neural network, Γ(Z i，p ) is the Gaussian kernel function; is the adjustment error of the angular velocity of the UAV, is the output signal of the filter. In addition, the adaptive parameter The update law and interference effectiveness evaluation function h(θ i,p,2 ) and auxiliary function μ i,p1 The calculation expression is:

[0150]

[0151]

[0152] Among them, r i,p1 、r i,p2 and l i is a positive constant; θ i,p,2 is the interference effectiveness evaluation index of unknown aggregate interference, and is calculated as follows:

[0153]

[0154] Furthermore, the filter design in an ideal input signal is:

[0155]

[0156] Among them, v i,p is a positive constant; α i,p (t) is the designed virtual control law, and its calculation expression is:

[0157]

[0158] Among them, c i,p,1 >1 is a positive constant; h(·) is the interference effectiveness evaluation function; μ i,p2 is an auxiliary function, and its calculation expression is:

[0159]

[0160] 2) After switching the working mode, s i,p =-1. The ideal input signal and virtual control law are designed as follows:

[0161]

[0162] The meanings of the symbols and the calculation formulas are the same as those in 1).

[0163] First, the stability of the designed control signal and virtual control law in case 1) without reverse fault is proved. The filtering error is defined as:

[0164]

[0165] Then the derivative of (21) can be expressed as:

[0166]

[0167] Consider the following Lyapunov function:

[0168]

[0169] Taking the derivative of (39), we can obtain the following inequality:

[0170]

[0171] Further consider the following Lyapunov function:

[0172]

[0173] Taking the derivative of (41) and combining it with (40), we have:

[0174]

[0175] The stability analysis of the control signal and virtual controller after switching the working mode in case 2) is similar to the above.

[0176] Based on the above analysis, the designed swarm fault-tolerant control and interference utilization method based on the switching function method can ensure the stability of the UAV.

[0177] In this embodiment, by designing a set of working modes and a switching function, fault-tolerant control and interference utilization of a swarm of drones with unknown time-varying fault directions are realized, which effectively solves the problems of excessive reverse input and system state jitter in the traditional Nussbaum function method when solving the problem of unknown time-varying fault directions. This improves the transient performance of the drone and reduces the energy consumption of the system.

[0178] It should be noted that, in the present invention patent, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0179] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for fault-tolerant control and interference utilization of drone clustering, characterized in that: The following steps are involved: S1. Establish a dynamic model of the UAV attitude system with unknown time-varying fault direction; S2, designing an estimator to estimate the attitude angle output of the leader and generate a reference signal; S3. Design the working mode and switching function of the drone attitude system; specifically include the following contents: According to the dynamic model and fault model of the UAV attitude system, there are 8 potential fault direction modes. The designed working mode set encoded in binary mode with {-1,1} as elements is: Where diag{·} represents a diagonal matrix; Based on the above working mode, the following switching function is designed for each drone: Among them, S i (t) = diag{s i,1 ,s i,2 ,s i,3 } represents the working mode of drone i; bin2dec represents binary to decimal conversion; sign represents the sign function; αi(t) and Represents auxiliary functions; p=1,2,3 is the event triggering time, and its judgment conditions are: Among them, the event condition g i,p (t) is calculated as: g i,p (t)=|z i,p (t)|-β(t) Among them, z i,p represents the adjustment error, β(t) is the set event monitoring boundary, and the function is expressed as: Among them, ε0, ε ∞ and is a positive constant and satisfies ε0>ε ∞ ; S4. Design interference observer and interference effectiveness evaluation index for mismatched interference; S5. Design the ideal control signal for input soft saturation and UAV attitude system.

2. A method for fault-tolerant control and interference utilization of drone clustering according to claim 1, characterized in that: The UAV attitude system described in S1 includes a follower UAV attitude system with unknown time-varying fault direction and a leader UAV attitude system with unknown time-varying fault direction; The function of the attitude system of the following UAV with unknown time-varying fault direction is expressed as: yi=xi, p,1， i=1,...,N Among them, x i,1,1 、x i,2,1 、x i,3,1 Respectively represent the roll angle, pitch angle and yaw angle of the UAV attitude system; x i,1,2 、x i,2,2 、x i,3,2 Represent the angular velocities of roll, pitch and yaw respectively; represents the state vector; d i,p,1 and d i,p,2 denote the unmatched disturbance and matched disturbance with bounded derivatives respectively; y i,p Represents the Euler angle output of the drone; f i,p A nonlinear term representing the attitude of the drone; is the torque under the UAV fault, which is expressed as the system input: Among them, u i,p represents normal torque input; ρ i,p represents the failure factor, satisfying ρ i,p ≠0; r i,p Indicates additive fault; The nonlinear term of the above UAV attitude is specifically expressed as: Among them, J x , J y , J z , J r Represents the moment of inertia and rotor inertia in the x, y, and z directions of its own coordinates; I p Indicates air resistance parameters, p = 1, 2, 3; Ω r represents the differential angular velocity of the rotor; the communication topology of the followers in the drone cluster contains a directed spanning tree and the root node can obtain the signal of the leader; The function of the leader UAV attitude system with unknown time-varying fault direction is expressed as: in, and They represent the leader’s Euler angle and the corresponding angular velocity respectively; l i Indicates the number of the group to which drone i belongs; They represent the output, input and nonlinear term of the leader UAV respectively; the nonlinear term of the leader UAV has the same form as that of the follower UAV; Assume that the derivative of the leader UAV output is bounded, expressed as in Is a normal number.

3. A method for fault-tolerant control and interference utilization of drone clustering according to claim 2, characterized in that: The S2 specifically includes the following contents: The communication interaction between drones is used to design an estimator for each follower drone to estimate the leader's information and achieve group consistency of the output signals of each estimator. The specific function is expressed as: in, represents the output of the estimator; Represents the input signal of the observer, and its function is expressed as: Where c represents a positive constant; represents the output vector of the leader drone and its derivative is ω i represents the group consistency error; and Respectively represent group l i The set of drones that have signal transmission and reception links with other drones in the group; represents the neighbor set of drone i; b i Represents the traction gain, which has b for the traction node i =1, otherwise b i =0; follow the root node and collection of the drone topology graph and The nodes in a are pulling nodes; ij with a ik Represents the topological weight between UAVs.

4. The method for fault-tolerant control and interference utilization of drone clustering according to claim 3 is characterized in that: The S4 specifically includes the following contents: Design a disturbance observer, specifically: in, represents the observed output of the mismatched disturbance, ψ i,p,1 represents the auxiliary function, φ i,p,1 >1 is a constant; Based on the above interference observer design, the interference effectiveness evaluation index is further designed. The specific contents are as follows: Among them, θ i,p,1 Represents the interference effectiveness evaluation index, when θ i,p,1 >0, it indicates that the disturbance is harmful to the completion of the control target and full compensation is performed; when θ i,p,1 When <0, it indicates that the interference is beneficial to the completion of the control target, and faster convergence and lower energy consumption can be achieved by utilizing the interference; e i,p1 It represents the conversion value of the adjustment error, and its calculation expression is: Among them, z i,p Indicates adjustment error; and k i,p They represent the conversion function and the safety distance respectively, and their calculation expressions are: in, δ and a i,p is a positive constant, δ≥2.

5. The method for fault-tolerant control and interference utilization of drone clustering according to claim 4 is characterized in that: The S5 specifically includes the following contents: Based on the working mode and switching function designed in S3, the input signal of the drone is expressed as: in, It represents the saturation input implemented at the software level, and its calculation expression is: Among them, u sat represents a normal number, Represents the ideal input signal of the design, and its calculation expressions are respectively: 1) No fault condition, at this time s i,p =1, p = 1, 2, 3; the ideal input signal design is: Among them, c i,p,2 Constant, c i,p,2 >1; Z i =[x i,1,1 ,x i,1,2 ,x i,2,1 ,x i,2,2 ,x i,3,1 ,x i,3,2 ] T represents the input of the neural network; Γ(Zi, p) represents the Gaussian kernel function; Indicates the adjustment error of the drone’s angular velocity; represents the output signal of the filter; represents the adaptive parameter, The update law and interference effectiveness evaluation function h(θ i,p,2 ) and auxiliary function μ i,p1 The calculation expression is: Among them, r i,p1 、r i,p2 and l i is a positive constant; θ i,p,2 It represents the interference effectiveness evaluation index of unknown aggregate interference, and its calculation method is: The filter design for an ideal input signal is: Among them, v i,p is a positive constant; α i,p (t) is the designed virtual control law, and its calculation expression is: Among them, c i,p,1 >1 is a positive constant; h(·) represents the interference effectiveness evaluation function; μ i,p2 Represents an auxiliary function, whose calculation expression is: 2) After switching the working mode, s i,p =0; the ideal input signal and virtual control law are designed as follows: The meaning and calculation formula of each symbol in the above formula are the same as the corresponding symbols in step 1).

Images